The proposed study uses data and specimens from a Southwest Oncology Group (SWOG) breast cancer clinical trial of CAF or CMF to investigate inherited factors that may impact prognosis. Disease-free survival following adjuvant therapy, as well as treatment-related toxicity, are likely predicted by a number patient and tumor characteristics. Pharmacogenetics, inherited differences in metabolism of chemotherapeutic agents and variability in capabilities to protect ceils (tumor and normal) from treatment-generated reactive oxygen species (ROS), may be of particular importance, but has been understudied, to date. For example, cyclophosphamide undergoes hepatic activation by CYP3A4 and metabolites may be detoxified by conjugation with glutathione, catalyzed by glutathione S-transferases P1 and A1. We hypothesize that determination of ultimate cytotoxic metabolites by these variants will impact treatment efficacy. Furthermore, variability in enzymes that protect the mitochondria (MnSOD, GPX1, CAT), key to controlling cancer cell death, and those in the secondary line of defense against products of oxidative stress (GSTM1, GSTT1), may also impact disease-free survival after treatment for breast cancer. These hypotheses are supported by our pilot data, in which women with GSTP1 and GSTA1 variants encoding lower enzyme activity and those with gene deletions in GSTM1 and GSTT1 had significantly less risk of recurrence and death after treatment for breast cancer. Lesser protection of cells from cytotoxic and reactive species could also result in more severe toxicity. We propose analysis of systemic, inherited DNA derived from normal archived tissue for polymorphisms in CYP3A4, GSTP1, GSTA1, MnSOD, GPX1, CAT, GSTM1 and GSTT1 among a subset of women who participated in a SWOG clinical trial in relation to recurrence and toxicity. Eligibility criteria included T1-3a node-negative invasive adenocarcinomas of the breast with tumors < 2cm and positive for estrogen and progesterone receptors. Based on flow cytometry results (S-phase), women were assigned to a higher risk group (n=528) and were randomized to either CMF or CAF, + or - tamoxifen. The lower risk group received no adjuvant treatment. DNA from these women will be genotyped and results merged with the SWOG patient database. Kaplan-Meier survival function, log rank tests for survival differences and Cox proportional hazards models will be used to determine the impact of genetic variability on disease-free survival, adjusting for other prognostic factors. Regression analysis will be used for associations between genotypes and toxicity endpoints. To evaluate if genetic polymorphisms have an association with survival that is independent of treatment, analyses will be conducted among women who received no adjuvant therapy. Findings from this study may be useful in prediction of which patients will respond to treatment for breast cancer and which will not, with an effort towards individualized therapy. Individualized dosing according to genotype could contribute to increased efficacy and decreased toxicity among patients being treated for cancer.